Upstream research in genomics is increasingly an information science: de novo DNA synthesis from an information specification (i.e., the DNA sequence) is becoming cheaper at a rate even faster than Moore's law might predict. Multi-gene and multi-allele diagnostics assemble together many different informational components and thus resemble the types of complex products produced by the information technology (IT) industries. As genomics research grows increasingly information-based, it risks facing some of the peculiar intellectual property problems that have plagued the IT field. However, if lessons from IT's experience can be applied to genomics, this could prevent emerging genomics research from unnecessarily encountering the same hurdles, and allow it to better reap the benefits of new processes and technologies. Project 1 will be dedicated to addressing the unique and important intellectual property challenges posed by genomics as information, and applying lessons learned from IT's distinctive struggles in this area. Over the last 10 years, the IT industries have complained vigorously about problems with the low quality of IT patents. Specifically, they argue that IT patents are often overly broad - covering obvious technology widely used in the industry or claiming territory that the researcher has not actually invented -.- and that patents have vague boundaries that fail to tell researchers when they are impinging on other patents. In addition, IT industries have been concerned that because their complex products implicate hundreds if not thousands of patents, patent thickets may impede their ability to develop such products. The quality-related issue that may be most salient for bench researchers is the complaint that IT patients regularly evade patent law's disclosure requirement. In order to receive a patent, patent holders are required to describe their invention in sufficient detail to allow others in the field to understand and improve upon (or invent around) what the patent holder claims. The disclosure bargain promotes progress by allowing the underlying information to be available to the public even during the life of the patent In the IT field, there is increasing evidence that patent holders do not describe technology adequately for researchers to have access to this underlying information. As genomics becomes more like IT, a similar problem could stall the progress of genomic research. In fact, the effects of low quality patents could be more pronounced in genomics than in IT. When patents were extended to IT products, IT was already a robust industry, so the patents did not extend to foundational technologies. In nascent areas of genomic research, however, there is a risk that patents could cover foundational technologies and basic information inputs, further slowing growth and innovation. Project 1 will use empirical and theoretical modeling techniques to determine the extent to which the kinds of problems associated with patents on information in the IT industries are likely to emerge in genomics. It will focus particularly on those technologies and applications where a primary output of genomics research is information that may be subject to intellectual property rights. The original empirical research for the project will focus on questions of patent quality (thus addressing whether low quality patents are proliferating). At the same time, it will be drawing on related empirical research on possible patent thickets in multi-gene/multi-allele diagnostics and full-genome DNA sequencing (Project 3), while other portions of it are being pursued by Dr. Chandrasekharan under a recently funded ROS grant. Project 1 will also explore the effect that some recent administrative and judicial efforts to address perceived problems with patents in IT are likely to have on the validity and enforceability of genomic patents that claim information, most notably DNA sequences (e.g., a nucleic acid sequence encoding a protein with amino acid sequence SEQ ID XXX). Finally, Project 1 will turn from patent law to another area of intellectual property, by examining hurdles that copyright law and publishing practices pose to the full scientific exploitation of genomics as information. For example, copyright law and publishing practices make it difficult to get full text access, which is necessary in order to use sophisticated algorithms to explore publications, and to integrate those publications with the wealth of other genomic data and studies being produced. The question being pursued here is, if genomics is rapidly becoming a pure information science, to what extent can we learn from the successes of the World Wide Web to create a science system that seamlessly merges published research with the underlying genomic data on which that research relies?